Supervised LLE in ICA space for facial expression recognition

Qijun Zhao, Dapeng Zhang, Hongtao Lu

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

27 Citations (Scopus)

Abstract

Locally linear embedding (LLE) is an unsupervised nonlinear manifold learning algorithm. It performs well in visualizing data yet has a very poor recognition rate in facial expression recognition. In this paper, to improve the performance of LLE in facial expression recognition, we first employ the independent component analysis (ICA) technique to preprocess the face images such that they are represented by some independent components and some noise is filtered from them. We then propose a Supervised LLE (SLLE) algorithm to learn the hidden manifold. SLLE constructs the neighborhood graphs for the data according to the Euclidean distances between them and the cluster information of them. Its embedding step is the same as that of LLE. Finally, we use a generalized regression neural network (GRNN) to learn the implicit nonlinear mapping from the ICA space to the embedded manifold. Experiments on the JAFFE database show promising results.
Original languageEnglish
Title of host publicationProceedings of 2005 International Conference on Neural Networks and Brain Proceedings, ICNNB'05
Pages1970-1975
Number of pages6
Volume3
Publication statusPublished - 1 Dec 2005
Event2005 International Conference on Neural Networks and Brain Proceedings, ICNNB'05 - Beijing, China
Duration: 13 Oct 200515 Oct 2005

Conference

Conference2005 International Conference on Neural Networks and Brain Proceedings, ICNNB'05
Country/TerritoryChina
CityBeijing
Period13/10/0515/10/05

ASJC Scopus subject areas

  • Engineering(all)

Cite this